Formaldehyde (HCHO), sodium molybdate dihydrate (Na₂MoO₄ 2H₂O), thiourea (NH₂CSNH₂), N,N-dimethylformamide (C₃H₇NO), ethanol absolute (C₂H₆O), and isopropanol (C₃H₈O) were bought from Sinopharm Chemical Reagent Co., Ltd (Beijing, China), and 5, 5-dimethyl-1-pyrroline-N-oxide (DMPO) used for radical analysis was provided by Sigma Chemical Co. Ltd. Ordered mesoporous carbon (OMC) was provided by Low-Dimension Materials. The type of OMC was CMK-3 and provided by the Low-Dimension Materials. The specific surface area was 1,000 m²/g, total pore volume was 1.35 cm³/g, and the micropore volume was 0.01 cm³/g. All chemicals used in the experiments were analytically pure and provided by the company without further purification.

The MoS₂/OMC heterojunction photocatalyst was successfully synthesized by a hydrothermal process (Bian et al., 2012). Sodium molybdate dihydrate and thiourea were used as precursors to prepare the MoS₂/OMC photocatalyst. A total of 100 mg OMC was added to C₃H₇NO solution and then mixed the solution with Na₂MoO₄ 2H₂O (0.15 g) and NH₂CSNH₂ (0.24 g) by ultrasonication; the mixed solution was then transferred to a stainless steel high-pressure reactor, followed by the hydrothermal reaction at 200°C for 24 h. Finally, the black precipitate was collected by centrifugation after naturally cooling to room temperature and washed with deionized water and ethanol. Subsequently, the precipitate was dried at 80°C for 24 h under vacuum and calcined for 2 h, noting as MoS₂/OMC photocatalyst.

The pure MoS₂ materials were prepared as follows: Na₂MoO₄ 2H₂O (0.3 g) and NH₂CSNH₂ (0.8 g) were mixed well into distilled water (65 ml) by ultrasonication and then the mixed solution was transferred to a stainless steel high-pressure reactor, followed by the hydrothermal reaction at 200°C for 24 h. The black precipitate was collected by centrifugation after naturally cooling to room temperature and washed with deionized water and ethanol. The precipitate was dried at 80°C for 24 h under the vacuum and calcined for 2 h, and after that, the prepared powders were further grinded into smaller powders with a mortar.

The crystal phase structure and mesoporous order degree of photocatalysts were analyzed by using an X-ray diffraction (XRD) instrument (Bruker, D8 ADVANCE). The scanning speed was 0.05°/s, and the scanning range of 2θ was 10°–90°. The microscopic morphology of the material was characterized by scanning electron microscopy (SEM) (S-4800). The microstructure of the photocatalysts was further analyzed with a transmission electron microscope (TEM, Phillips model CM200). The physical adsorption properties of the materials were measured by a Micromeritics ASAP 2010 adsorption apparatus. Besides, photocurrent analysis was performed with an electrochemical workstation (CHI 660D).

The top of double-layer glass reactor (300 ml) was sealed by a quartz sheet, and the outer interlayer was refluxing water in the photocatalytic reaction process to ensure the photocatalytic reaction was carried out at room temperature. The excitation light source used in the reaction was a 300 W xenon lamp, supported with a UV420 filter (PLS-SXE300D, Beijing), ensuring the excitation light source in the process of the photocatalytic reaction was visible light (420 nm≤λ ≤ 780 nm). The photocatalyst powder was uniformly dispersed in a petri dish (diameter 6 cm) containing 1 g of ethanol absolute under ultrasonication and then dried at 40°C.

The petri dishes loaded with photocatalysts were placed at the inner bottom of the double-layer glass reactor, and the visible light source excited by the xenon lamp was illuminated to the surface of the petri dishes through the quartz window at the top. The distance between the petri dishes and the UV420 filter was 10 cm. The reactor was purged with 60 ml/min of high-purity nitrogen for 45 min to eliminate CO₂, eliminating VOCs and other gases in the reactor and gas pipeline before the reaction. During the photocatalytic reaction, a gas-tight syringe (SGE, 500 μL) was used to collect reaction gas at intervals through the acquisition window of the double-layer glass reactor, and the reaction gas was detected by using a gas chromatograph. The removal efficiency (η) of formaldehyde under visible light irradiation was calculated by the following equation: η = C 0 − C t C 0 × 100 %, (1) where C_t and C₀ are the concentrations of formaldehyde in the reactor at each given time and initial time, respectively.

The morphology of photocatalysts (MoS₂, OMC, and MoS₂/OMC) was analyzed by SEM and XRD. As shown in Figure 1A, many hollow flowers structure could be observed for the pure MoS₂ photocatalysts. As shown in Figures 1A,B, the micron-scale three-dimensional structure of OMC materials contained many pores. According to the definition of the International Union of Pure and Applied Chemistry based on pore width (usually pore diameter or layer distance of slit shape pores) (Han et al., 2019), pores in the materials can be classified into macropores (pore size >50 nm), mesopores (2 nm < pore size <50 nm), and micropores (pore size <2 nm). The pore size of OMC was 5.57 nm, and lots of mesopores were present in the OMC materials. Therefore, the pore structure of OMC had a larger specific surface area and more edge active sites (Zhang et al., 2021c). As shown in Figure 1C, the flower structure of MoS₂ was successfully assembled into the surface of OMC. The particular structure contributed to the full contact between the degradable compounds and the catalytic active sites, which not only maintained the high activity of photocatalysts but also avoided agglomeration (Ismael, 2021). Corresponding elemental mapping analysis is shown in Figures 1D–G; the homogeneous distribution of C, O, Mo, and S elements can be observed for MoS₂/OMC. Besides, the XRD spectra of MoS₂ and MoS₂/OMC are shown in Figure 1H. Comparing with the standard card, the corresponding crystal plane of the diffraction peak was (002), (100), (103), and (110), in accordance with the hexagonal structure of MoS₂ (Wei et al., 2022). The peak of MoS₂/OMC was observed at 2θ = 24.8° on the (002) crystal surface of OMC (JCPDS75-1621), indicating that the mesoporous carbon in the complex was amorphous carbon material after hydrothermal treatment. In order to better demonstrate the heterojunction microstructure, the TEM technique was used to analyze the surface morphology of MoS₂/OMC photocatalysts. As shown in Figures 1I,J, MoS₂/OMC showed an ordered nanoflower structure composed of ultrathin nanosheets with an expanded interlayer spacing, with an interplanar spacing of 0.82 nm indexed as (002) plane of MoS₂.

BET analysis and the hole diameter distribution of MoS₂/OMC photocatalyst were evaluated. The N₂ adsorption isotherms of the MoS₂/OMC photocatalyst was type IV based on IUPAC classification, showing the existence of both micro- and mesopores (Figure 2A) in accordance with the result of pore size distributions (Figure 2B). The introduction of MoS₂ would affect the microstructures of OMC and change the pore textures of OMC. Before modification, the specific surface area for OMC was 1,000 m²/g and the total pore volume was 1.35 cm³/g, while the calculated specific surface area of the MoS₂/OMC photocatalyst was 613 m²/g after MoS₂ decoration. As a result, MoS₂ decoration for preparing the MoS₂/OMC photocatalyst might block the mesopores of OMC to some extent. MoS₂ attached on the surface of OMC could hinder the agglomeration of MoS₂/OMC with a structure with thinner nanosheets and smaller size. Mesoporous carbon materials had a good surface chemical inert stability and mechanical strength, and their high specific surface area and uniform pores could also provide abundant reaction sites for improving the photocatalytic performance (Qiu et al., 2021).

For the pure MoS₂ sample, the onset edge of the absorption peak is about at 760 nm, corresponding to a bandgap of 1.63 eV (Zhou et al., 2013). Therefore, it was urgent to prepare the micro-heterojunction structure to improve the photocatalytic activity. Photocurrent spectroscopy was used to detect the generation and separation of photogenerated electrons. As shown in Figure 3, compared with pure MoS₂, the photocurrent density of the MoS₂/OMC was increased, showing that the introduction of OMC increased the separation rate of photogenerated electron–hole. The photocurrent result confirmed that the heterostructure promoted the separation of photo-induced charges efficiently and improved the photocatalytic performance.

The photocatalytic performances of different photocatalysts (MoS₂, OMC, and MoS₂/OMC) were analyzed. As shown in Figure 4, MoS₂/OMC showed the obviously increased photocatalytic performances under visible light irradiation (λ > 420 nm) in comparison to the pure MoS₂ and OMC (the removal efficiency of formaldehyde for MoS₂, OMC, and MoS₂/OMC were 30.59, 82.34, and 5.54%). The modified MoS₂/OMC photocatalysts effectively inhibited the recombination of photogenerated electrons and photogenerated holes and prolonged the life of photogenerated carriers. Therefore, the heterojunction microstructure of MoS₂/OMC photocatalysts was a benefit for increasing the removal efficiency of formaldehyde.

The effect of initial concentration on formaldehyde removal was shown in Figure 5A, the removal efficiency of formaldehyde was improved from 67.72 to 84.34% when the initial concentration on formaldehyde was increased from 0.5 to 1.0 mg/m³. The result indicated that the active sites of the MoS₂/OMC surface were not fully utilized with the lower concentration of formaldehyde, and the higher concentration of formaldehyde was benefited for increasing the utilization rate of MoS₂/OMC. When the initial concentration of formaldehyde was low, the catalytic behavior of photocatalysts was mainly affected by the mass transfer and diffusion from the main gas phase to the surface of photocatalysts. The increasing initial concentration of formaldehyde increased the partial pressure and thus accelerated the reaction rate. But, the removal efficiency of formaldehyde was reduced when the formaldehyde concentration increased from 1.0 to 2.5 mg/m³; therefore, the optimal formaldehyde concentration for formaldehyde removal using MoS₂/OMC as photocatalysts was 1.0 mg/m³. The mass transfer diffusion from the main body of the gas phase to the surface of the catalyst accelerated the formaldehyde adsorbed by the active site on the MoS₂/OMC photocatalyst surface when the formaldehyde concentration was higher. Meanwhile, the effects of mass transfer diffusion and adsorption of formaldehyde were obviously weakened, and the catalytic reaction of formaldehyde on the surface of MoS₂/OMC photocatalysts had become the main factor affecting the reaction.

The amount of photocatalysts was an important factor to determine the formaldehyde removal. In this experiment, the effect of photocatalysts amount on formaldehyde removal is shown in Figure 5B. The removal efficiency of formaldehyde improved gradually when the photocatalyst amount was increased from 0.01 to 0.03 g/m³. So the increasing amount of photocatalysts could increase the active site of the reaction, which could promote the catalytic reaction. However, the removal efficiency of formaldehyde reduced when the photocatalyst amount was increased from 0.03 to 0.05 g/m³. The further increase in the photocatalyst amount inhibited the removal efficiency of formaldehyde improving due to the agglomeration phenomenon and low mass transfer efficiency.

Radical quenching experiments were further conducted to confirm the generated radicals responsible for formaldehyde removal. Isopropanol (IPA) worked as the OH radical quencher, and N₂ gas was used to reduce the O₂ · ^- radicals (Zhang et al., 2017). As shown in Figure 6, the removal efficiency of formaldehyde was decreased to 62.04% when adding IPA to the reaction solution. Furthermore, when N₂ was bubbled into the reaction solution, the removal efficiency of formaldehyde was even decreased to 65.84%. The results indicated that both OH and O₂ ^- radicals were the major active radical species for formaldehyde removal in the photocatalytic process. The photocatalytic mechanism for formaldehyde removal using MoS₂/OMC photocatalysts was explained. During the photocatalytic process, the photogenerated electrons would react with the adsorbed O₂ molecules on the surface of MoS₂/OMC photocatalysts to produce O₂ · ^- , and the photogenerated holes would react with the adsorbed H₂O molecules on the surface to generate OH (Lan et al., 2018). These reactive oxygen species would further react and oxidize the adsorbed HCHO molecules on the surface into CO₂ and H₂O.

Mesoporous OMC materials had a high specific surface area, and uniform pores could also provide abundant reaction sites for improving the photocatalytic permanence. Besides, the MoS₂ hollow flowers uniformly grew on the surface of OMC through the hydrothermal process, reducing the agglomeration for MoS₂/OMC photocatalysts and enhancing the photocatalytic permanence.

The stability of the MoS₂/OMC photocatalyst was an important index to evaluate its practical value, and the stability performance of MoS₂/OMC materials is shown in Figure 7A. The degradation efficiency of formaldehyde still remained over 90%, and there was no obvious reduction after six cycles of testing, which proved that the MoS₂/OMC photocatalyst could keep its high mechanical strength, good stability, and easy to recycle. This was mainly because the hollow spherical MoS₂ was assembled into an orderly structure of OMC. This structure of MoS₂/OMC materials not only maintained high activity and high specific surface area of the nanosheets but also greatly increased the number of edge active sites. At the same time, the structure was stable and easy to recycle. Therefore, it effectively overcame the defect of agglomeration of nanomaterials. The photocatalytic reactor containing MoS₂/OMC photocatalyst was highly active under the optimum operating conditions (formaldehyde concentration for formaldehyde removal was 1.0 mg/m³, and photocatalysts amount was 0.03 g/m³). To evaluate the photocatalytic performance of MoS₂/OMC, the photodegradation of HCHO into CO₂ and H₂O under visible light irradiation was employed. As shown in Figure 7B, the amount of the degraded HCHO was almost the same as generated CO₂. The results showed that the MoS₂/OMC photocatalyst had high activity and good stability.